But rather to ensure the experimental methodology was sound, it cites the appropriate previous work, that it is significant enough, etc.
I actually think that calculation verification and statistical significance verification is a great idea, but shouldn't be done by "peers", but rather someone who specializes only in that. More of a type of auditor.
The article made waves because it didn't surprise anyone, because in fact a ton of household staples are full of toxins (spices are full of lead, etc).
So the reason it made waves is because it confirmed everyone's previous biases. Which is why people defend it even when that "confirmation" was based on an error of simple arithmetic.
Reminds me of a bunch of studies that were taken out by Excel errors, like that economics failure about national debt going higher than GDP being some sort of trigger for societal failure. The people who boosted that study were the people who were saying the same thing, with no study, for no other reason than that the numbers are of a similar magnitude. After it was shown that the study showed no such thing, its supporters insisted that it was still true because of course it was true.
https://stanfordreview.org/clarifying-the-implications-of-th...
Your comment is the first time I've ever heard anything about lead in spices, and I read a lot of news. In my experience, very few things are actually "common knowledge", it just seems that way when everyone in our particular social circles knows something.
Personally, I would expect a lot of products to be near a lot of limits for various toxins. Especially around food, my expectation is for limits to be safe with a nice-sized buffer, sometimes too big of a buffer. The point of having limits is that products inside them are fine.
Most papers are full of errors nobody ever catches (or cares to catch).
Science is about making reality the arbiter of truth. What is true is what you can demonstrate in an experiment pretty much anywhere. Science is self-correcting not because other scientists take a look at your work before it gets published, but because a theory that is not backed up by reality will be harder to build on, make worse predictions, and generally not be useful for actual work. It doesn't matter how much "hype" built up around LK-99 for example, because you could not build anything out of it, because it didn't work as claimed, and no amount of belief can overcome that.
Science is self correcting because the universe is not convinced by rhetoric, and will only ever follow it's own rules.
Tellingly, scientists who have been around for a while do not consider a published paper as especially meaningful. They do not consider peer review as vouching for a paper's correctness. They have seen hundreds of papers pass peer review and make it into Science and Nature and turn out to be bunk. No scientific process will be perfect, so some amount of papers should be expected to be wrong, no matter how genuinely careful and perfect the scientists who wrote it were.
It's really sad how over-indexed the lay-public has gotten on peer review. It's just not how science works. It can still be somewhat useful for more popular journals who have to pick and choose what to publish anyway, but it's only useful if the peer reviewers are themselves domain experts in a rigorous and formalized scientific discipline.